5buyai × AIFUNS
推广
← 返回资讯动态
Anthropic
@AnthropicAI
We're an AI safety and research company that builds reliable, interpretable, and steerable AI systems. Talk to our AI assistant @claudeai on https://t.co/FhDI3KQh0n.
1.6M关注者
2正在关注
20本站收录
2含视频
Amazon Web Services Amazon Web Services Lightreel AI Lightreel AI ElevenLabs Developers ElevenLabs Developers Figma Figma Netflix Netflix DigitalOcean DigitalOcean Notion Notion Runway Runway Spline Spline Microsoft Azure Microsoft Azure YouTube YouTube Telegram Messenger Telegram Messenger Google Ads Google Ads PixVerse PixVerse LovartAI LovartAI OpenClaw🦞 OpenClaw🦞 Canva Canva MiniMax (official) MiniMax (official) GitHub GitHub Spotify Spotify Red Hat Red Hat MiniMax Design (H3) MiniMax Design (H3) X X Google Cloud Google Cloud Instagram Instagram Pika Pika Cursor Cursor OpenAI OpenAI Twitch Twitch Midjourney Midjourney
全部 🖼 图文 🎬 视频
Could a model one day align its stronger successors? As a first test, we had Sonnet 5 post-train an early checkpoint of Opus 4.8, a more capable model. It reached safety scores approaching those of production Opus 4.8, which went through our full alignment training.
Across 10 alignment failures, Claude reliably improved safety scores without degrading capabilities. Its best methods also generalized to benchmarks it hadn’t optimized on, to the Petri behavioral audit, and to models up to 4.7x larger.
Claude “hill-climbed” safety benchmarks for common misalignments like deception or sycophancy, with one constraint: it had to preserve general capabilities. We then tested its best methods on held-out benchmarks to see if they'd generalize.
MHS currently best covers lab and manufacturing equipment. Many developers are already using Claude Code to operate hardware like boards and cameras; our research preview will help us extend MHS to these devices, so they can all work under one interface.
There’s more to learn before we open source MHS. LLMs still lack physical intuition, having learned about the physical world from text and images. The research preview will let us build more safety evaluations and strengthen protections for using AI in the physical world.
In early testing, AI agents used MHS to: Run a drug-discovery experiment with real-time error handling at Genentech Compress an imaging experiment from weeks to a day at HHMI Janelia Research Campus Improve laser stabilization on QuEra's quantum computers from 58% to 99.3%
Connecting AI to hardware requires days or weeks of bespoke integration, with no standard way for agents to operate equipment safely. MHS cuts integration to hours or minutes, provides an interface that makes devices discoverable, and enables agents to operate them safely.
Now, we want to scale this research model. If you're a researcher and would like access to our tools to pursue work you can't otherwise do today, we’d like to hear from you. You can express interest here: https://forms.gle/rmLjTvibven9CmDFA
The other two studies are ongoing: HIP Lab is studying how Claude's behavior relates to how people feel when using AI, while METR is estimating real-world productivity gains from coding agents. We'll share more from both soon.
Three research groups—Stanford’s Social and Language Technologies lab, Oxford’s Human Information Processing Lab, and METR—designed independent studies to analyze the aggregated outputs from 250,000 https://Claude.ai or Claude Code conversations between April and May 2026.
One of our highest priorities remains launching an access program for scientists to use our most capable models. We expect to share more on this soon. Opus 5 remains our most capable model available for life science research.
以上内容聚合自 X 公开时间线,版权归原作者所有,点击可查看原文。本站与该账号无隶属关系。
💬 需要帮忙吗?
购买 / 支付 / 账号问题,点这里问客服 →

公告

①敬告:本站( 5buyai X AIFUNS )为人工发货(UTC+8的9:00~21:00),你购买的账号将通过邮件发送到你的下单邮箱,网站不会存储你的账号信息,请下单后查看你的下单邮箱,如你填错下单邮箱,请及时联系我们!

②关于支付:如果你无法完成付款或者遇到了订单已经支付,却没有成功跳转,查询订单时显示未支付;可以点击右下角聊天部件与5buyai.com 团队聊天,认准支付商家“宇柒云阁

③订阅我们的消息:Telegram 通知群

➤➤免责声明:本站提供的商品仅供学习和测试使用,请勿将本站资料用于任何违反当地法律法规的行为。

➤➤警告:下单付款后,无法退款。